AI Paper Atlas

Research map / Label-efficient and robust learning / Semi-supervised learning

PAF: Perturbation-Aware Filtering for Open-Set Semi-Supervised Learning

Yinan Han, Qing-Yuan Jiang

CVPR 2026 · Label-efficient and robust learning › Semi-supervised learning · approach: Confidence-based pseudo-label selection

Read the paper

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add perturbation-aware filtering of unlabeled data (ablation)PAF · CIFAR-100 (mismatch 0.3)seen-class accuracy (%): 71.2 -> 77.7Improved
use separate projection heads for consistency and rotation branches (ablation)PAF · CIFAR-100 (mismatch 0.3)seen-class accuracy (%): 61.8 -> 77.7Improved
add representation consistency loss (ablation)PAF · CIFAR-100 (mismatch 0.3)seen-class accuracy (%): 75.1 -> 77.7Improved
add auxiliary rotation prediction loss (ablation)PAF · CIFAR-100 (mismatch 0.3)seen-class accuracy (%): 75.6 -> 77.7Improved
use representation instability instead of confidence mutation for filtering (ablation)PAF · CIFAR-100 (mismatch 0.3)seen-class accuracy (%): 74.5 -> 77.7Improved
use representation instability instead of confidence mutation for filtering (ablation)PAF · CIFAR-100 (mismatch 0.3)OOD AUC (%): 78.2 -> 84.5Improved
use representation-level filter instead of confidence-based filter (ablation)PAF · CIFAR-100 (mismatch 0.3)ID sample retention rate and OOD filtering rate: +9.9% ID retention, +12.2% OOD filteringImproved
increase perturbation trials Ka beyond 1 (ablation)PAF · CIFAR-100 (mismatch 0.3)AUC (%)No clear effect
use Otsu adaptive threshold instead of fixed or percentile thresholds (ablation)PAF · CIFAR-100 (mismatch 0.3)seen-class accuracy and AUCImproved
rescale Otsu threshold by gamma around 1.0 (ablation)PAF · CIFAR-100 (mismatch 0.3)seen-class accuracy and AUCNo clear effect
plug PAF into CFSG-CLIPCFSG-CLIP · CUB-200-2011 (fine-grained OSSL)seen-class accuracy (%): 84.73 -> 85.16Improved
use PAF instead of T2T for open-set SSL training costPAF · CIFAR-100 (mismatch 0.3)training time (hours): 29.3 -> 23.1Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 4 of 4 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy